Description

In certain versions of Oracle ZFS Storage Appliance Software, the Gen 4 Logzilla(TM) Solid State drives (Oracle PN: 7045627, Model number: Z16IZF2EUSUN73G) can have increased writes to the SSD eventually causing them to fail with a SMART trip for exceeding the endurance threshold.

Note: Self-Monitoring, Analysis and Reporting Technology (SMART) errors are a near-term prediction of drive failure.

Occurrence

This issue can occur on the following Oracle ZFS Storage Appliance Software versions:

for the following platforms with Gen 4 Logzilla SSDs:

• Sun Disk Shelf
• Oracle Exalogic  Elastic Cloud X3-2
• SPARC Super Cluster T4-4
• Sun ZFS Storage 7120
• Sun ZFS Storage 7320
• Sun ZFS Storage 7420

Notes:

1. Some of the affected systems listed above were initially shipped with Gen 1 LogZilla SSDs (not Gen 4) and are therefore not affected by this issue.

2. If a system is running any of the above versions, it is highly recommended to take action as soon as possible - either upgrade to versions that have the fix or use the "Appliance Workflow" Workaround to make the necessary changes (see Workaround/Resolution section below and <Document:1595313.1>).

3. No other platforms (7110, 7210, 7310, 7410, Simulator) or Oracle ZFS Storage Appliance releases are affected by this issue.

The Gen 4 SSD model number can be determined from the display in the BUI:



To determine the Oracle ZFS Storage Appliance Software version on the device, see the following screen from the BUI:

 

3. To see which systems are supported by each release, see the current releases and previous releases tables in <Document:2021771.1>

Symptoms

The drive will fail with a SMART trip, 0x80 with additional sense key 0x4c for exceeding the endurance threshold. When the Logzilla gets failed by SMART, the ZFS
appliance gives the following fault:

      "SMART health-monitoring firmware reported that a failure is imminent on disk <DISK>"

Where <DISK> will be in the range DISK 20 - DISK 23

This fault can be seen under "maintenance problems" screen in either the BUI or CLI.

The following is an example of the fault displayed in the BUI:



This information can also be viewed from the CLI as in the following:



The following is an example of the errors in the support bundle:

    Sep 24 2013 09:08:37.567569961 ereport.io.scsi.disk.predictive-failure
    nvlist version: 0
    class = ereport.io.scsi.disk.predictive-failure
    version = 0x0
    ena = 0xf6439be584d06001
    detector = (embedded nvlist)
    nvlist version: 0
        version = 0x0
        scheme = hc
        hc-root =
        serial = STM0001766FE
        part = STEC-Z16IZF2EUSUN73G    ----->  Gen 4 Logzilla solid state drive model #
        revision = 9410
        authority = (embedded nvlist)
    nvlist version: 0
        product-id = SUN-Storage-J4410
        product-sn = 1311FMM00U
        server-id =
        chassis-id = 1313FMD01C
    (end authority)
    hc-list = (array of embedded nvlists)
    (start hc-list[0])
    nvlist version: 0
        hc-name = ses-enclosure
        hc-id = 0
        (end hc-list[0])
        (start hc-list[1])
        nvlist version: 0
            hc-name = bay
            hc-id = 20
        (end hc-list[1])
        (start hc-list[2])
        nvlist version: 0
            hc-name = disk
            hc-id = 0
        (end hc-list[2])
    (end detector)
    additional-sense-code = 0x80            ------ SMART ERROR with 0x80
    additional-sense-code-qualifier = 0x4c  ------ With additional Sense qualifier - 0x4c
    __ttl = 0x1
    __tod = 0x5241aaf5 0x21d46e29

Workaround

This issue is addressed in the following Oracle ZFS Storage Appliance Software releases:

It is highly recommended to upgrade as soon as possible to one of the released versions which has this fix.

For the following platforms with Gen 4 Logzilla SSDs:

• Sun Disk Shelf
• Sun ZFS Storage 7120
• Sun ZFS Storage 7320
• Sun ZFS Storage 7420
• Oracle Exalogic  Elastic Cloud X3-2

Upgrade to the following:

For SPARC Super Cluster T4-4, upgrade to the following:

Note: If you are unable to  upgrade the system to the appropriate software version listed above, use the "Appliance Workflow" Workaround that makes the necessary changes, described in Problem <Document:1595313.1>, which outlines instructions on obtaining the "Appliance Workflow" Workaround and how to install and execute it. These Workaround changes will be persistent.

History

30-Oct-2013: Document released, issue is Resolved

Only the systems stated are affected and the others are not because :

1) Storage 7x10 platform only shipped Gen 1 SSDs, which are not impacted.
2) Gen4 SSDs were supported on 7x20 only from  versions Ak 1.4.0 and above.

This was fixed during early testing during NPI, so there were no
SRs nor escalations.

Questions regarding this document should be
addressed to sunalertpublication_us_grp@oracle.com
and copy the responsible engineer/submitter below.

Internal Contributor/Submitter:  Nachi.Nachiappan@oracle.com, kimito.sakata@oracle.com
Internal Eng Responsible Engineer: Nachi.Nachiappan@oracle.com
Internal Knowledge Analyst: david.mariotto@oracle.com
Internal Eng Business Unit Group:  Storage

References

This solution has no attachment